SavedAdd your journal

Journals / Afyon Kocatepe Üniversitesi Fen Bilimleri Dergisi / 2003 / Cilt: 3 - Sayı: 1-2

Uniform population and uniform operator in genetic algorithms

Genetik algoritmalarda düzenli populasyon ve düzenli operatör

Ali KARCI
Bilal ALATAŞ
Pages
11–26
Publication Date
2022-10-11
DOI
—
🏛️Bibliographic metadata and full text access for this article are provided via DergiPark.
DergiPark Record ↗
✨

AI Summary & Key Takeaways

AI Insight
⚡ TL;DR(Too Long; Didn't Read)

By using this operator with uniform population method, problems of genetic search are coped with and global solution is rapidly reached.

📌 3 Key Takeaways

🎯
Objective & Hypothesis:The main idea of the Genetic Algorithms (GA) are based on natural selection and genetic rules which is first to generate randomly solution sets for the problem to solve and then to start the genetic process to improve these solutions.
📊
Key Findings & Data:In this study, a new operator, developed by being inspired from the uniform population method, is proposed to eliminate the disadvantages of the randomly generated initial population in the genetic process and to improve the performance of the genetic search.
💡
Conclusion & Impact:By using this operator with uniform population method, problems of genetic search are coped with and global solution is rapidly reached.
🔬Design:Research Article
📍Scope:Comprehensive Sample
⚙️Variables:Domain Parameters

Abstract

The main idea of the Genetic Algorithms (GA) are based on natural selection and genetic rules which is first to generate randomly solution sets for the problem to solve and then to start the genetic process to improve these solutions. In this study, a new operator, developed by being inspired from the uniform population method, is proposed to eliminate the disadvantages of the randomly generated initial population in the genetic process and to improve the performance of the genetic search. By this operator, based on the problem, four or more offsprings of high quality from two best individuals are generated; genetic diversity is ensured and early convergence is prevented. By using this operator with uniform population method, problems of genetic search are coped with and global solution is rapidly reached.

Özet

Doğal seçim ve genetik kurallara dayandırılmış bir arama işlemi olan genetik algoritmaların (GA) temel mantığında başlangıçta çözülecek problem için gelişi güzel olarak çözümler kümesinin üretilmesi ve daha sonra bu çözümlerin genetik bir süreçle iyileştirilmesi vardır. Bu çalışmada, genetik süreç içerisinde gelişigüzel oluşturulan başlangıç populasyonun dezavantajlı durumunun giderilmesi ve genetik aramanın performansının arttırılması için düzenli populasyon yönteminden esinlenerek geliştirilen yeni bir operatör sunulmuştur. Bu operatörle iki ebeveynden dört ya da probleme özgü olarak daha fazla, kaliteli çocuk birey oluşturulmakta; genetik çeşitlilik sağlanmakta ve erken yakınsama önlenmektedir. Bu operatör ve düzenli populasyonla, genetik aramanın sözü edilen problemlerin üstesinden geldiği ve global çözüme daha kısa sürede ulaştığı görülmüştür.

🔖

Cite This Article

KARCI, A., ALATAŞ, B. (2022). Uniform population and uniform operator in genetic algorithms. Afyon Kocatepe Üniversitesi Fen Bilimleri Dergisi, Cilt: 3 - Sayı: 1-2, 11–26. https://ajindex.com/articles/429540

References (12)

  1. Srivastava R. P., Goldberg D. E., Verification of the Theory of Genetic Algorithm Continuation, Proc. of the Genetic and Evolutionary Computation Conference, San Fransisco, California, 623-630, (2001)
  2. Vasconcelos J. A., Ramirez J. A., Takahashi R. H. C., Saldanha R. R., Improvement in Genetic Algorithms, IEEE Trans. On Magnetics, 37: (3), 3414- 3417, (2001)
  3. Rowe J. E., PcHee N. F., The Effect of Crossover and Mutation Operators on Variable Length Linear Structures, Proc. of the Genetic and Evolutionary Computation Conference, San Fransisco, California, 535-542, (2001)
  4. Schumacher C., Vose M. D.L., Whitley D., The No Free Lunch and Problem Description Length, Proc. of the Genetic and Evolutionary Computation Conference, San Fransisco, California, 565-570, (2001)
  5. Branke J., Reducing the Sampling Variance When Searching for Robust Solutions, Proc. of the Genetic and Evolutionary Computation Conference, San Fransisco, California, 235-242, (2001)
  6. Cao Y., Wu Q. H., Study of Initial Population in Evolutionary Programming, Proc. of the European Control Conference, (1997)
  7. Hill R. R., A Monte Carlo Study of Genetic Algorithm Initial Population Generation Methods, Proc. of the 1999 Winter Simulation Conference, 543- 547, (1999)
  8. Karcı A., Arslan A., Bidirectional Evolutionary Heuristic for the Minimum Vertex-Cover Problem, Journal of Computer and Electrical Engineering, Elsevier, 29, 111-120, (2003)
  9. Karcı A., Çınar A., Comparison of Uniform Distributed Initial Population Method and Random Initial Population Method in Genetic Search, The 15th International Symposium on Computer and Information Sciences, 159-166, İstanbul, (2000)
  10. Karcı A., Arslan A., Genetik Algoritmalarda Düzenli Populasyon, GAP IV. MÜHENDİSLİK KONGRESİ, 601-607, (2002)
  11. Karcı, A. ve Arslan, A., Uniform Population in Genetic Algorithms İ.Ü. Journal of Electrical & Electronics, 2: (2), 495-504, (2002)
  12. Karcı A., Arslan A., Düzenli Populasyon (DP) Yönteminin Kaliteli Populasyon Üretme Yeteneğinin Olasılıksal Analizi, Politeknik Dergisi, Gazi Ünv., Teknik Eğt. Fak., Ankara, (Basımda).